Asymmetric Sacroiliac Joint Anatomy in Partial Lumbosacral Transitional Variations: Potential Impact on Clinical Testing in Sacral Dysfunctions T

Home , Lumbosacral joint, Sacroiliac joint

Medical Hypotheses 124 (2019) 110–113

Contents lists available at ScienceDirect

Medical Hypotheses

journal homepage: www.elsevier.com/locate/mehy

Asymmetric sacroiliac joint anatomy in partial lumbosacral transitional variations: Potential impact on clinical testing in sacral dysfunctions T

Niladri Kumar Mahatoa,b a Department of Pre-Clinical Sciences, Faculty of Medical Sciences, The University of the West Indies, St. Augustine, Trinidad and Tobago b Former Post-Doctoral Researcher, Ohio Musculoskeletal & Neurological Institute (OMNI), Department of Biomedical Sciences, Ohio University, Athens 45701, OH, USA

ARTICLE INFO ABSTRACT

Keywords: Lumbosacral transitional vertebrae (LSTV) anomalies may present as bio-mechanical dysfunctions leading to low Auricular surface back pain (LBP). Unilateral or incomplete/partial LSTVs have been documented to be associated with signiﬁcant Low back pain sacroiliac joint (SIJ) joint asymmetries. Objective evaluation of outcomes from routine clinical testing for sacral LSTV dysfunctions on these subsets of LSTV cannot be found in the literature. Based on quantitative studies available Sacral-dysfunction on LSTV-associated anatomical variations at the SIJ, this study hypothesizes probable outcomes of standard Sacral torsion palpatory clinical tests used to evaluate sacral dysfunctions in unilateral LSTV anomalies. Since LSTV is reported in a sizeable percentage in the general population and due to its proposed etiological relationship with LBP, these entities warrant attention in terms of the anatomical bases of related clinical assessments and their outcomes, as proposed in this hypothesis.

Introduction sacral dysfunctions [18–20]. However, outcomes of clinical dysfunction testing (for sacral torsions) in LSTV anomalies have neither been hy- Lumbosacral transitional variations (LSTV) are commonly detected pothesized or documented, especially in partial LSTVs that in most in clinical practice (Fig. 1) [1–3]. The prevalence of LSTV in the po- instances, present with (i) morphological asymmetries of the osseous pulation is estimated to be around 12.5% (range = 4.0%-35.9%), with sacroiliac articulations, (ii) facet tropism, and/or (iii) variations in joint alterations in lumbar spine curvature, disc degeneration, peripheral ligamentous attachments [3,13,21–23]. Although there are different nerve entrapment, LBP and SIJ dysfunctions being reported with these anatomical variants observed across the LSTV spectrum, the objective anatomical variations at the lumbosacral junction [4,5]. A subset of of this hypothesis is to discuss variants that are partial/unilateral and these variations such as unilateral L5-S1 accessory articulations (Cas- result in asymmetric morphological changes at the sacroiliac articular tellvi Type II A), partial/incomplete/unilateral sacralisation (partial areas. Accordingly, this study hypothesizes how these specific mor- fusion of L5 vertebra/Castellvi Type III A), partial/incomplete/uni- phometric changes may impact clinical testing in LSTV-associated lateral lumbarization (separation of S1segment), or mixed Type IV lumbosacral and sacroiliac (sacral) dysfunctions. The hypothesis pre- anomalies have been characterized by Castellvi and other investigators sented in this study attempts to elucidate the possible relationship be- as asymmetric anomalies [6–9]. Such unilateral LSTVs at the lumbo- tween unilateral LSTVs and sacral dysfunctions and attempts to predict sacral junctions are also associated with ipsilateral asymmetries of outcomes of common clinical tests that are applied to diagnose such auricular surface dimension at the sacroiliac joint and facet mor- dysfunctions. For the purpose of facilitating the understanding of the phology. Partial LSTVs usually show up as significantly larger dimen- hypothetical concept, hypothesis for partial/incomplete sacralization sions of the auricular surface area and height on the side of the lum- and lumbarization, and the hypothesis for unilateral accessory L5-S1 bosacral fusion when compared to the unfused side [6,10,11]. articulations have been presented separately. Accordingly, significant variability in the sacral articular dimensions The PubMed database was searched for publications specifically may affect load bearing equilibrium and dynamics at the sacroiliac reporting quantitative, morphometric studies on sacral dimensions that joints [11–14]. As such, these variations in weight bearing may result in compared sacroiliac joint surfaces in the normal and LSTV associated back pain, unilateral sacral dysfunctions that can be diagnosed with lumbosacral junctions. Reports on the mechanisms of clinical testing in clinical dysfunction testing [1,15–17]. Signs and outcomes of palpatory sacral dysfunctions, sacroiliac asymmetries, associated biomechanical clinical evaluations have been well documented in lumbosacral and alterations and concomitant joint dysfunctions were reviewed from the

E-mail address: [email protected]. https://doi.org/10.1016/j.mehy.2019.02.002 Received 30 October 2018; Accepted 1 February 2019 0306-9877/ © 2019 Elsevier Ltd. All rights reserved. N.K. Mahato Medical Hypotheses 124 (2019) 110–113

Fig. 1. Castellvi et al. four subtype LSTV classiﬁcation. Type I includes unilateral (IA) or bilateral (IB) dysplastic transverse processes (19 mm wide, ≥cranio-caudal). Type II incomplete unilateral (II A) or bilateral (II B) ‘accessory’ diarthrodial articulation between an enlarged transverse process of the preceding lumbar vertebra and the sacral ala. Type III LSTV, unilateral (III A) or bilateral (III B) . Type IV was described as a mixed feature of Type II transition on one side and a Type III on the other . This classiﬁcation does not classify sacralization (6-segment sacrum) or lumbarisation (5-segment sacrum) as two dustinct morphological entities.

Table 1 The table shows a representative transitional state, the left-sided incomplete Sacralisation/Lumbarization. A left-sided incomplete fusion with smaller left auricular surface (column one) with common types of dysfunctions (column two), the predicted outcomes of routine clinical tests (column three to eight). L = Left; R = Right; LSTV = Lumbosacral Transitional Vertebrae; ILA = Infro-lateral Angle; LUF = Left Unilateral Flexion; LUF = Left Unilateral Flexion; LUE = Left Unilateral Extension; SS = Sacral Sulcus; ↑=in greater degrees than assessed in normal sacrum for the same dysfunction; =↓in lesser degrees than assessed in normal sacrum for the same dysfunction due to osseo-ligamentous asymmetry in unilateral LSTV.

Transitional Anomaly Problem Dysfunctional Seated Deep Base/ ILA Spring Test Comments LSTV Flexion Test Sacral Sulcus

Incomplete Sacralisation/Lumbarization (e.g. Left Unilateral (left) LUF L L L −ve More common than RUF sided asymmetry with smaller auricular surface Flexion or Extension LUE L R R +ve (Left) More common than RUE on the left/unaﬀected side) Bilateral Flexion or Bilateral F L > R L L −ve – Extension Bilateral E L > R ––+ve (> on SS, ILA comparatively the Left) shallow on the left Torsion In Flexion LonL ↑R ↓R ↑L −ve Less common than R on R (forward) RonR ↓L ↑L ↓R −ve More common form Torsion In Extension LonR ↑L ↓R shallow ↑L +ve More common form (backward) RonL ↓R ↑L shallow R↓ +ve Less common than R on L

Table 2 The table shows a representative transitional state, a left-sided unilateral L5-S1 accessory articulation (ﬁrst column) with slight left side-bend (left sacral axis), with common types of dysfunctions (column two), the predicted outcomes of routine clinical tests in these anomalies (column three to eight). L = Left; R = Right; LSTV = Lumbosacral Transitional Vertebrae; ILA = Infro-lateral Angle; LUF = Left Unilateral Flexion; LUF = Left Unilateral Flexion; LUE = Left Unilateral Extension; SS = Sacral Sulcus; ↑=in greater degrees than expected in normal sacrum for the same dysfunction; =↓in lesser degrees than expected in normal sacrum for the same dysfunction due to osseo-ligamentous asymmetry in unilateral LSTV. The study hypothesizes L5 rotation to be minimum in these LSTVs, thereby the predominant axis for torsions becoming the L5 side-bend. The corresponding outcomes are italicized.

Transitional Anomaly Problem Dysfunction Diagnosis Seated Deep Base/ ILA Spring Test Comments in LSTV Flexion Test Sacral Sulcus

Left-sided L5- S1 Accessory Articulation (e.g. Unilateral Flexion or RUF R R R −ve More common than LUF Left sided asymmetry with larger Extension RUE R L R +ve (Right) More common than LUE ligaments on the left/affected side) Bilateral Flexion or Bilateral F R > L R R −ve – Extension Bilateral E R > L ––+ve (> on SS, ILA comparatively the Right) shallow on the right Torsion In Flexion Left Rotation L on L ↑R ↑RL−ve L on L more common (forward) Left side- RonR L L ↑R −ve bend Right LonL ↑RR L−ve L on L more common Rotation RonR L L ↑R −ve Torsion In Extension Left Rotation LonR R ↑RL+ve R on L more common (backward) Left side- RonL ↑R L shallow R↑ +ve (Right) bend Right LonR R ↑RL+ve R on L more common Rotation RonL ↑R L shallow R↑ +ve (Right) literature. The outcomes of these clinical tests were reviewed in the Hypothesis and discussion light of the anatomical asymmetries at the SIJ and a set of hypotheses was formulated that could explain mechanistic relationships between A. Unilateral sacralisation/lumbarization (Castellvi Type IIIA): In unilateral LSTV anomalies and potential outcomes of sacral dysfunc- unilateral sacralisation/lumbarization, the dimension of the osseous SIJ tions being tested. The hypotheses related to the clinical testing are has been found to be smaller on the unfused side [11]. Additionally, discussed below. The hypothetical outcomes of the tests have been studies have shown the sacroiliac, ilio-lumbar and lumbosacral liga- compared with normal non-LSTV lumbosacral junctions and sacroiliac ments on the unaffected, non-LSTV side to be smaller compared to the joints[Tables 1 and 2] and figures (Figs. 2 and 3). Considering the LSTV-affected side in the same individual, or when compared to the limitations of this hypothesis-only study, investigative palpatory testing dimensions in the general population. These morphological alterations data from normal or LSTV-related spines per se, have not been included may have a cause-effect relationship related to the hypermobility ob- in this project. served on the contralateral, non-LSTV affected joint. This may eventually present as dysfunction and pain on the non-LSTV side [24,25]. Some studies have reported the pain to originate from the non-LSTV side at the initiation of LBP. Possibly, the unaffected side could be more

111 N.K. Mahato Medical Hypotheses 124 (2019) 110–113

(ii) The Sacral Sulcus (SS): In general, the palpatory depth of the sacral sulcus in sacroiliac dysfunctions can be predicted to be greater or lesser depending on the type and side of dysfunction. However, with partial LSTV, relatively greater laxity at the non-LSTV side may change the palpatory position of the sacral sulcus. Since the side of the LSTV presents a larger sacral auricular area, the ipsilateral joint may be more stable due to the fusion of the L5 ele- ment. The SS, therefore, may present deeper on the joint ipsilateral to the LSTV. (iii) The Infro-lateral Angle (ILA): This test is carried out in the prone position where the examiner palpates the sacral sulcus and the inferior angle of the sacrum on each side. The sacral sulci and inferior angles are assessed to detect if they are symmetrical or asymmetrical. Forward and backward sacral torsions are assessed by noting the excursions of the sacral sulci and the inferior angles Fig. 2. The schema explaining the anatomical basis of the current hypotheses. having patient move up onto his/her elbows in the same prone Regardless of the increased dynamicity of the non-LSTV side and/or the left/ position. The side with the smaller SIJ and smaller, less extensive right presentation of the dysfunction, any associated pain may originate from either/both sides. ligamentous attachments may demonstrate larger excursions of the ILA compared to the contralateral side depending on the type of dysfunction. Therefore, SIJ with strong ligamentous attachments (non-LSTV) may demonstrate smaller ILAs and vice versa. (iv) The Spring Test (ST): In torsional dysfunctions, the axis of rotation identified by palpatory compressive techniques over the pronated sacrum is usually represented by the side with the larger auricular surface area (Fig. 3). The side of the SIJ with larger articular surface is more stable and may force a large swinging arc toward the opposite joint. Additionally, the Spring test may yield greater manual displacement over a smaller, asymmetric, SIJ in extension and in backward torsional dysfunctions. Therefore, the LSTV-re- Fig. 3. Representative images from the two unilateral LSTVs discussed in this lated SIJ may represent the rotatory axis of the torsional dys- hypothesis. A. Partial/Unilateral sacralisation showing the circled LSTV af- function, whereas the contralateral, non-LSTV side may demon- fected side. The opposite non-LSTV side is relatively unstable allowing greater strate greater displacements with the ST. flexion–extension or torsional excursions . The hypothesis predicts the torsional axis (dashed line) to be ipsilateral to the LSTV allowing backward-forward B. Unilateral L5-S1 accessory articulation (Castellvi Type IIA): The torsional dysfunction ( curved arrow) predominantly on the non-LSTV side. B. L5-S1 Accessory Articulation (circled) . The opposite non-LSTV side may be common Castellvi Type IIA anomaly presents a cartilaginous di-ar- relatively unstable allowing greater flexion–extension or torsional excursion . throdial articulation between one of the transverse processes of the L5 The hypothesis proposes the torsional axis (dashed line) to be ipsilateral to the vertebra with the sacral alae. Unlike partial sacralization, the accessory side of the LSTV that permits backward-forward torsional dysfunction ( curved articulation does not fuse with the sacral ala to augment the ipsilateral arrow) predominantly on the unaffected, contralateral, non-LSTV side. auricular surface area [13]. The lumbosacral and ilio-lumbar ligaments Regardless of the increased dynamicity of the non-LSTV side and/or the left/ demonstrate strong attachments on the ipsilateral, LSTV affected side. right presentation of the dysfunction, reported pain may originate from eother Consequentially, the movement at the LSTV-affected SIJ may become side. Pain in situation such as in B is usually reported at the site of the accessory restricted and at times, painful [28–30]. These changes may influence articulation ( open asterisk ). outcomes of side-bending and rotation tests. The predicted outcomes [Table 2] and salient features for the tests are described next. The ex- ‘vulnerable’ to dysfunctions earlier than the ipsilateral LSTV side due to ample used in the hypothesis assumes a left-sided Type IIA LSTV that this hypermobility. Therefore, the non-LSTV side may be the side pre- forces the L5 (in vivo) to take a slightly left side-bent position (sacral senting with greater degrees of flexion–extension or torsional excur- axis shifted to the left). LSTV literature also indicates that these sions if not stabilized by outgrowth of osteophytes, as evidenced by anomalies are associated with facet asymmetries [31,32]. The facets some studies [26,27]. The predicted outcomes of specific tests have closer to the accessory articulations are found to be smaller (even ru- been tabulated [Table 1] and features of the hypothesis related to dy- dimentary), allowing a greater arc of rotation at the contralateral, non- namic clinical examination are enumerated below: LSTV SIJ. Therefore, in our example we will consider the L5 to be fixed and rotated slightly to the left. Based on these conditions, the outcomes (i) Seated Flexion Tests (SFT): SFT or the seated forward flexion test is of clinical tests may be explained in a left accessory-articulated, partial used to assess decreased motion at one of the sacroiliac joints LSTV. performed in the seated position. The examiner has a hand on each of the patient’s posterior superior iliac spines (PSIS). Next, the (i) Seated Flexion Tests (SFT): SFTs may be prominent on the non- patient bends forward while the examiner compares the movement LSTV side due to the potential hypermobility predicted due to the of each PSIS. A test is positive when greater superior motion is felt absence of greater osseo-ligamentous restrainers on this side. on one PSIS compared to the other. SFTs may be prominently Accordingly, although pain may be reported from the side bearing elicited on the unaffected, non-LSTV side of the SIJ due to the the accessory LSTV articulation, paradoxically, the dysfunction potential hypermobility resulting from the contralateral fusion. may be detected on the contralateral, non-LSTV SIJ due to the Accordingly, in a LSTV-associated back pain patient, dysfunction resultant hypermobility. may be more readily detected on the normal side in the pelvis. In (ii) The Sacral Sulcus (SS): The depth of the sacral sulcus may be cases of a bilateral flexion–extension or torsional dysfunctions, the greater or shallower depending on the type and side of dysfunc- SFT values may be greater on the non-LSTV side; or when com- tion. However, the sacral articulation contralateral to the accessory pared to a similar dysfunction in a patient without such LSTV. LSTV may demonstrate greater mobility for palpatory signs for

112 N.K. Mahato Medical Hypotheses 124 (2019) 110–113

dysfunction e.g., the depth of the of the sacral sulcus or the posi- doi.org/10.1016/j.mehy.2019.02.002. tion of the sacral base, since the SIJ ipsilateral to the L5-S1 accessory articulation is usually stable and less mobile. References (iii) The Infro-Lateral Angle (ILA): The side contralateral to the LSTV presents less intensive ligamentous attachments compared to the [1] French HD, et al. Lumbosacral transitional vertebrae and its prevalence in the more stable ipsilateral side with the LSTV. Such anatomical dis- Australian population. Global Spine J 2014;4(4):229–32. [2] Apazidis A, et al. The prevalence of transitional vertebrae in the lumbar spine. Spine crepancy may feature larger excursions of the ILA at the con- J 2011;11(9):858–62. tralateral, non-LSTV side, depending upon the type of dysfunction [3] Tins BJ, Balain B. Incidence of numerical variants and transitional lumbosacral detected. vertebrae on whole-spine MRI. Insights Imaging 2016;7(2):199–203. [4] Konin GP, Walz DM. Lumbosacral transitional vertebrae: classification, imaging (iv) The Spring Test (ST): In torsional dysfunctions in L5-S1 accessory findings, and clinical relevance. AJNR Am J Neuroradiol 2010;31(10):1778–86. articulation LSTVs, the axis of rotation is may be represented by [5] Jancuska JM, Spivak JM, Bendo JA. A review of symptomatic lumbosacral transi- the side affected by the LSTV anomaly. This may occur secondary tional vertebrae: Bertolotti's syndrome. Int J Spine Surg 2015;9:42. to the ‘fixation’ of the joints with stronger and abundant osseo- [6] Tague RG. High assimilation of the sacrum in a sample of American skeletons: prevalence, pelvic size, and obstetrical and evolutionary implications. Am J Phys ligamentous connections. Therefore, the SIJ on the LSTV side may Anthropol 2009;138(4):429–38. be more stable. This side may act as the axis of rotation/torsion of [7] Castellvi AE, Goldstein LA, Chan DP. Lumbosacral transitional vertebrae and their the sacrum and thus, determine the rotational orientation of the relationship with lumbar extradural defects. Spine (Phila Pa 1976) 1984;9(5):493–5. lumbosacral joint. This assumption is reinforced by observations in [8] Mahato NK. Implications of structural variations in the human sacrum: why is an the literature on the facet tropism and lumbosacral movement data anatomical classification crucial? Surg Radiol Anat 2016;38(8):947–54. fi fi that indicate minimization of ipsilateral sacroiliac rotational mo- [9] Mahato NK. Rede ning lumbosacral transitional vertebrae (LSTV) classi cation: integrating the full spectrum of morphological alterations in a biomechanical con- tion at lumbosacral junctions with partial/incomplete LSTV tinuum. Med Hypotheses 2013;81(1):76–81. [21,22,27]. Accordingly, the ST may yield greater manual dis- [10] Miller AN, Routt MLC. Variations in sacral morphology and implications for ilio- placement over the normal, non-LSTV side in extension and sacral screw fixation. J Am Acad Orthopaedic Surg 2012;20(1):8–16. [11] Mahato NK. Variable positions of the sacral auricular surface: classification and backward torsional dysfunctions with asymmetric, unilateral LSTV importance. Neurosurg Focus 2010;28(3). fusions. [12] Connolly LP, et al. Skeletal scintigraphy of young patients with low-back pain and a lumbosacral transitional vertebra. J Nucl Med 2003;44(6):909–14. [13] Illeez OG, et al. The transitional vertebra and sacroiliac joint dysfunction associa- Consequences of the hypothesis and discussion tion. Eur Spine J 2018;27(1):187–93. [14] Yang SW, Langrana NA, Lee CK. Biomechanics of lumbosacral spinal fusion in Morphological and biomechanical data available from studies ob- combined compression-torsion loads. Spine (Phila Pa 1976) 1986;11(9):937–41. jectively point out potential etiological implications of LSTV in LBP. [15] Montgomery DM, Egan IF, Pollard HP. Palpable unilateral sacral prominence as a clinical sign of lower limb anisomelia: a pilot study. J Manipulative Physiol Ther Consequently, correct interpretation of the clinical assessment out- 1995;18(6):353–6. comes of sacral dysfunctions associated with LSTV patients may help in [16] Pal GP. Mechanism of production of scoliosis. A hypothesis. Spine (Phila Pa 1976) – accurate diagnosis and clinical decision-making in such cases [3,9,31]. 1991;16(3):288 92. [17] Pal GP. Weight transmission through the sacrum in man. J Anat 1989;162:9–17. Since a large percentage of LSTV may eventually manifest as lumbo- [18] Fortin JD, Ballard KE. The frequency of accessory sacroiliac joints. Clin Anat sacral disc degeneration and SIJ dysfunction, clinical (and radiological) 2009;22:876–7. fi detection LSTV anomalies may be helpful in customizing specific LSTV [19] Forst SL, et al. The sacroiliac joint: anatomy, physiology and clinical signi cance. Pain Phys 2006;9(1):61–7. related management. This hypothesis proposes the evaluation of me- [20] Mitchell TD, et al. The predictive value of the sacral base pressure test in detecting chanistic links and the interpretation of clinical tests results in uni- specific types of sacroiliac dysfunction. J Chiropr Med 2007;6(2):45–55. lateral LSTV. As postulated in this hypothesis, the oucomes of these [21] Vleeming A, et al. The sacroiliac joint: an overview of its anatomy, function and – fi potential clinical implications. J Anat 2012;221(6):537 67. tests may appear paradoxical in terms of unexpected xation of the [22] Snijders CJ, Vleeming A, Stoeckart R. Transfer of lumbosacral load to iliac bones sacral axis, relative right-left motion discrepancies in the unilateral and legs Part 1: biomechanics of self-bracing of the sacroiliac joints and its sig- sacralization / lumbarization scenario. The observations presented on nificance for treatment and exercise. Clin Biomech (Bristol, Avon) fi 1993;8(6):285–94. common LSTV asymmetries may help to develop speci c manual [23] Mahato NK. Facet dimensions, orientation, and symmetry at L5–S1 junction in techniques to detect partial/unilateral LSTV. The hypothetical differ- lumbosacral transitional states. Spine 2011;36(9):E569–73. ences presented in this study may help to explain bio-mechanistic re- [24] Schwarzer AC, et al. The false-positive rate of uncontrolled diagnostic blocks of the – lationship between complex LSTV anatomy, anomalous SIJ motion and lumbar zygapophysial joints. Pain 1994;58(2):195 200. [25] Schwarzer AC, et al. The relative contributions of the disc and zygapophyseal joint LBP. As a future direction, this hypothesis may be confirmed by in chronic low-back-pain. Spine 1994;19(7):801–6. studying back pain patients, with or without unilateral LSTV, pre- [26] Pal GP, Routal RV, Saggu SK. The orientation of the articular facets of the zyga- senting to clinics suggestive of SIJ dysfunction. A single, blinded ex- pophyseal joints at the cervical and upper thoracic region. J Anat 2001;198(Pt 4):431–41. aminer could perform all the tests described here across patients with [27] Goodwin RR, et al. Distraction and compression loads enhance spine torsional normal or LSTV associated SIJ anatomy. Given the current estimates of stiffness. J Biomech 1994;27(8):1049–57. [28] Muir JM. Chiropractic management of a patient with low back pain and Castellvi LSTV prevalence, the high predictive values of the select tests to detect – fi type II lumbosacral transitional vertebrae. J Chiropr Med 2012;11(4):254 9. sacral dysfunctions (speci city/sensitivity), and the high probability of [29] Adams MA. Biomechanics of back pain. Acupunct Med 2004;22(4):178–88. positive findings in dysfunction related LBP, a reasonable number of [30] Panjabi MM. The stabilizing system of the spine. Part I. Function, dysfunction, dysfunctional patient cohort would be required to demonstrate an ac- adaptation, and enhancement. J Spinal Disord 1992;5(4):383–9. discussion 397. [31] Li W, et al. Lumbar facet joint motion in patients with degenerative disc disease at ceptable odds ratio to satisfactorily correlate the predicted outcomes affected and adjacent levels: an in vivo biomechanical study. Spine (Phila Pa 1976) physical exam findings and the LSTV variants discussed in this hy- 2011;36(10):E629–37. pothesis. [32] Mahato NK. Association of Rudimentary sacral zygapophyseal facets and accessory fi and ligamentous articulations: implications for load transmission at the L5–S1 There was no nancial support involved in this work. junction. Clin Anat 2010;23(6):707–11.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://

113